Structured medical database generation method and system based on medical text information

Abstract

The present invention discloses a structured medical database generation method based on medical text information. The method comprises: obtaining input medical text information; determining a natural semantic processing model corresponding to the medical text information, and using the natural semantic processing model to carry out deep natural semantic analysis processing on the medical text information to obtain a processing result; and determining the medical variable corresponding to each processed data in the processing result, and inputting each processing data into the corresponding position of the corresponding medical variable to obtain a structured medical database. According to the method disclosed by the present invention, the automation degree and intelligence degree of the structured medical database generation can be improved, the labor cost can be greatly reduced, and efficiency of the structured medical database generation can be improved. The present invention also discloses a structured medical database generation system based on medical text information with the beneficial effects.

Description

technical field [0001] The invention relates to the technical field of medical data processing, in particular to a method and system for generating a structured medical database based on medical text information. Background technique [0002] The utilization of unstructured medical text information has always been a major difficulty in the technical field. In the prior art, medical text information is generally processed manually or semi-manually. Most doctors who need to use it or practitioners in related industries use manual reading of medical texts and standardized entry to process unstructured historical medical data (retrospective data). The general method is roughly that the electronic structured form (eCRF) is designed and programmed by the relevant personnel themselves or a third-party technology provider, and then the relevant personnel manually scan the text data with the naked eye one by one, and manually enter the relevant information into the structured form a...

AI Technical Summary

Problems solved by technology

However, there are some special cases, such as "lung cancer" and "small cell carcinoma of the lower right lung" (note that you can't find it if you use lung cancer as the keyword here), or when you face a non-standard expression of a cancer, such as abbreviations or typos Wait, the existing method cannot handle

The lack of these technologies leads to the failure of the accuracy and recall of the entire information extraction to achieve the expected results

That is to say, there are insufficient automation and intelligence in the processing process in the existing technology, and the labor cost is extremely high.

Images